- Email: [email protected]
ACC pathway activity and HSP70
Neuroscience Letters 453 (2009) 49–53
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Ketogenic diet attenuates kainic acid-induced hippocampal cell death by decreasing AMPK/ACC pathway activity and HSP70 Byeong Tak Jeon a , Dong Hoon Lee a , Kyu Hong Kim b , Hyun Joon Kim a , Sang Soo Kang a , Gyeong Jae Cho a , Wan Sung Choi a , Gu Seob Roh a,∗ a
Department of Anatomy and Neurobiology, Institute of Health Sciences, Medical Research Center for Neural Dysfunction, Gyeongsang National University, School of Medicine, 92 Chilam-dong, Jinju, Gyeongnam, 660-751, Republic of Korea Department of Neurosurgery, Masan Samgsung Hospital, Sungkyunkwan University School of Medicine, Masan, Gyeongnam, 630-723, Republic of Korea
b
a r t i c l e
i n f o
Article history: Received 11 November 2008 Received in revised form 28 January 2009 Accepted 28 January 2009 Keywords: Ketogenic diet Kainic acid AMPK ACC HSP70 Hippocampus
a b s t r a c t The ketogenic diet (KD) prevents kainic acid (KA)-induced hippocampal cell death. There are reports that AMP-activated protein kinase (AMPK) activation regulates the intracellular signaling pathways involved in cellular survival or apoptotic cell death. In this study, we investigated the effect of the KD consumption on the expression of signaling pathway proteins AMPK and ACC, and heat shock protein (HSP) 70 in mouse hippocampus after KA treatment. Mice were fed the KD for 6 weeks and then sacrificed 48 h after KA (30 mg/kg) injection. The marked cell death found commonly in normal diet (ND)-fed mice treated with KA was not observed in the KD-fed KA-treated mice. Western blot analysis revealed that phosphorylation of AMPK and ACC was increased after KA treatment. However, phosphorylation of these proteins was reduced in those animals that received the KD. In addition, increased expression of HSP70 in the hippocampus of KA-treated mice was decreased in animals receiving the KD. These results indicate that the KD promotes neuroprotective effects through suppression of the AMPK cascade and that HSP70 is involved in neuronal cell death or oxidative stress. © 2009 Elsevier Ireland Ltd. All rights reserved.
The ketogenic diet (KD) is a high-fat and low-carbohydrate diet primarily used to treat juvenile refractory epilepsy [1,25]. Despite its use in clinical therapy and experimental animal studies, it remains unclear what mechanisms underlie its seizure-suppressive action and neuroprotection. Like fasting, the KD has beneficial effects on epilepsy, results in limited glucose availability due to a very low carbohydrate intake, and forces the body to use ketones as an alternate source of acetyl-CoA to generate ATP [26,30]. During consumption of the KD, marked alterations in brain energy metabolism occur. Ketones partly replace glucose as fuel, increasing levels of ATP and purines which are critically involved in neuron–glia interactions, neuromodulation, and synaptic plasticity [10,18]. AMP-activated protein kinase (AMPK) acts as a multifunctional metabolic sensor in the brain [13,23]. AMPK is activated under conditions of glucose deprivation, heat shock, oxidative stress, and ischemia [4]. Once activated, AMPK suppresses the key enzymes involved in ATP-consuming anabolic pathways and increases cellular ATP supply [13]. AMPK stimulates fatty acid oxidation by phosphorylating and inhibiting acetyl-CoA carboxylase (ACC) [9]. AMPK regulates a variety of neuronal cell functions including
∗ Corresponding author. Tel.: +82 55 751 8735; fax: +82 55 759 0779. E-mail address: [email protected] (G.S. Roh). 0304-3940/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2009.01.068
survival, apoptosis, and brain metabolic plasticity [2,5,8]. Modest activation of AMPK induces neurogenesis and improved cognition in animals, but over-activation reduces cognition and increases neural apoptosis and mortality [6]. Under conditions of oxidative stress, heat shock protein (HSP) expression temporarily increases to assist recovery by enhancing the cell’s ability to cope with increased levels of denatured proteins [29]. HSP70, which is induced in areas damaged by KA-induced seizure, is a biochemical marker of neural injury and influences apoptotic cell death [15,27]. Neuronal cells need large amounts of energy, and abnormalities in energy metabolism are detrimental to neurophysiological functions. Previous reports have suggested that the KD increases ATP and the phosphocreatine-to-creatine ratio of energy reserves [7]. Since a seizure induces massive energy depletion and is followed by neuronal cell death, we hypothesized that preactivation of AMPK may itself play an important role in energy metabolism after seizure activity. The present study was undertaken to evaluate the expression of signal pathway proteins AMPK and ACC, and HSP70 in KA-induced hippocampal cell death following administration of the KD. Male ICR mice purchased from Japan SLC (Hamamatsu, Japan) were maintained in the animal facility of the Gyeongsang National University School of Medicine. Mice were treated in accordance
50
B.T. Jeon et al. / Neuroscience Letters 453 (2009) 49–53
with standard guidelines for laboratory animal care. Mice were fed ad libitum either a Harlan Teklad TD 96355 diet (KD group) or a standard rodent chow (normal diet, ND group) for 6 weeks, beginning at postnatal day 21. KD-fed mice were fasted overnight prior to initiation of the KD to facilitate the development of ketonemia [25]. After 6 weeks consuming a KD or ND, mice were treated with a subcutaneous injection of 30 mg/kg KA (Ascent Scientific, North Somerset, UK) emulsified in 0.9% normal saline. All animals were seizurenaïve when tested. Mice in control groups received 0.9% normal saline. Seizure behavior was monitored for 2 h after KA injection, and seizures were categorized into five grades (I–V), according to a previously defined scale [11]. Each experimental group (n = 10) was sacrificed at 48 h after KA treatment. For tissue analysis, mice (n = 4 per group) were perfused transcardially with heparinized saline followed by 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS). After 6 h post-fixation in the same fixative, the brains were sequentially immersed in 0.1 M PBS containing 15% sucrose and then in PBS containing 30% sucrose at 4 ◦ C, until they were completely submerged. The brains were cut into 40 m coronal sections and stained with cresyl violet, and the slides were evaluated by microscopy. Neuronal loss in the pyramidal layers of each section was scored on a 0–3 scale by observers blinded to the treatment conditions: 0, no lesion; 1, minimal lesion localized to CA3; 2, cellular loss in CA3 with some preservation of cellular architecture and some normal cellular components; 3, complete disruption of normal cellular architecture and components. To visualize degenerating neurons by Fluoro-Jade B (FJB) staining, slides containing brain sections were air-dried for 30 min, and then immersed in 1% sodium hydroxide in 80% ethanol for 5 min. The slides were then transferred to 70% ethanol and incubated for 2 min before transfer to deionized water (DW) for 2 min incubation, after which they were incubated in freshly prepared 0.06% KMnO4 for 10 min. Slides were then rinsed in DW for another 2 min before being incubated in 0.0004% FJB solution (Chemicon, CA, USA) for 20 min. The slides were rinsed three times in DW for 1 min, and excess water was removed by briefly hanging the slides vertically on a paper towel. They were then air dried for 10 min on a warmer, and immersed in xylene for 2 min before placing a coverslip with permount. The degenerating neurons were visualized as glowing green cells using a fluorescence microscope. The number of FJB-stained degenerating neurons was counted manually. For protein extraction, frozen mouse hippocampal samples (n = 6 per group) were transferred to sterile 1.5 mL microcentrifuge tubes containing 200 L lysis buffer (15 mM HEPES, pH 7.9, 0.25 M sucrose, 60 mM KCl, 10 mM NaCl, 1 mM EGTA, 1 mM PMSF, and 2 mM NaF). Homogenized tissues were incubated for 10 min on ice and then sonicated. Samples were then centrifuged at 4 ◦ C for 30 min at 12,000 rpm, and the supernatants were transferred to clean vials. Protein concentrations were determined using a Bio-Rad protein assay (Bio-Rad, USA), and samples were stored at −80 ◦ C. For phospho-AMPK (p-AMPK), AMPK, phospho-ACC (p-ACC), ACC, and HSP70 protein analysis, hippocampal lysates (30 g per lane) were separated by 8 or 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by electrophoretic transfer onto a polyvinylidene difluoride membrane (Millipore, MA, USA). The membranes were probed with antibodies against p-AMPK (Cell Signaling Technology, MA, USA), AMPK (Cell Signaling), p-ACC (Cell Signaling), ACC (Cell Signaling), or HSP70 (Santa Cruz Biotechnology, CA) and visualized using an ECL substrate (Pierce, Rockford, IL). To determine the neuroprotective effects of KD, median histopathologic score values were analyzed by nonparametric test (Mann–Whitney rank sum test). Differences between the groups (ND, ND+KA, KD+KA, and KD) were determined by one-way analysis of variance (ANOVA), followed by post hoc analysis using a Student–Newman–Keuls test. Values are expressed as the mean ± standard error of the mean (S.E.M.). A P value < 0.05 was considered statistically significant.
Fig. 1. Effect of KD on body weight and latency to seizure. (A) Ketogenic diet was begun at postnatal day 21 and was continued for 6 weeks. To promote ketosis induction, KD-fed mice group was deprived of food and water overnight prior to diet initiation. During the dietary treatment, body weight of ND-fed mice (n = 20) and KDfed mice (n = 20) showed no significant differences. (B) All mice were treated with KA (30 mg/kg, s.c.). In KD-fed mice, the latency to seizure onset was delayed compared to ND-fed mice. The data represent the mean ± S.E.M. of 20 mice. Significant (P < 0.05) difference compared with ND-fed mice.
KD-fed mice tolerated their diet well for 6 weeks. Other than the oily appearance of their fur, KD-fed mice showed no difference in behavior or health when compared with ND-fed mice. Apart from an initial weight loss in the ND- or KD-fed mice, mice in both groups experienced a similar weight gain within the time allotted (Fig. 1A). ND- and KD-fed mice showed increasing immobility that resulted in rigidity within 7 and 10 min (P = 0.01), respectively, after KA (30 mg/kg) injection (Fig. 1B). Although the latency of the seizures elicited by KA was retarded in KD-fed mice, the severity of the seizures was not significantly attenuated. During the 2 h post-injection observation period, all of the mice in both groups displayed head bobbing, circling behavior, and falling behavior, corresponding to grade IV. To determine whether KD plays a protective role in KA-induced hippocampal cell death, we performed cresyl violet and FJB staining (Fig. 2) of brain tissue sections. Cresyl violet staining revealed significant cell loss in the CA3 pyramidal layers of ND-fed KA-induced seizure mice, with pyramidal neurons appearing pyknotic (Fig. 2B). In contrast, cresyl violet staining in KD-fed KA-treated mice showed a significant level of protection against hippocampal cell death (Fig. 2C). Neuronal loss in ND+KA-fed mice (median score = 2.5) was significantly higher (P < 0.001) than that observed in KD+KA-fed mice (median score = 0.50). The FJB finding was consistent with cresyl violet staining (Fig. 2E and F). FJB-positive cells were observed in the CA3 region of ND-fed KA-induced seizure mice, whereas the signal was markedly reduced in KA-injected KD-fed mice (ND+KA-treated mice: 55.0 ± 2.74, KD+KA-treated mice: 7.5 ± 0.65; P < 0.001). No FJB-positive cells were observed in non-KA injected ND- or KD-fed mice.
B.T. Jeon et al. / Neuroscience Letters 453 (2009) 49–53
51
Fig. 2. Effect of KD on neuroprotection in the hippocampus of ND- or KD-fed mice after KA treatment. (A)–(D) Cresyl violet stained sections, showing the cytoarchitecture from (A) ND-, (B) ND+KA-, (C) KD+KA-, and (D) KD-treated mice. (E) and (F) Fluoro-Jade B stained section, showing the cytoarchitecture from (E) ND+KA- and (F) KD+KA-treated mice. Magnification: 400×. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
To explore whether phosphorylation of AMPK or ACC in the hippocampus of KA-treated mice is altered by KD, we performed Western blot analysis (Fig. 3). KA treatment resulted in an increased level of p-AMPK (P < 0.05) (Fig. 3A and B). However, KD attenuated the increase of AMPK activation in the hippocampus of KA-treated mice (P = 0.45). KD-fed mice showed a tendency toward increased AMPK activation over that of ND-fed mice; however, this
increase was not statistically significant (P = 0.059). In addition, a KA-induced increase in ACC phosphorylation was significantly inhibited by the KD (P < 0.05) (Fig. 3C and D). Consistent with results of AMPK activation, there was an increase in ACC phosphorylation in the hippocampus of KD-fed mice (P < 0.05). Finally, Western blot analysis was performed to determine the effect of KD on HSP70 expression under conditions of KA-induced hippocampal cell death
52
B.T. Jeon et al. / Neuroscience Letters 453 (2009) 49–53
Fig. 3. Effect of KD on the phosphorylation of AMPK and ACC in the hippocampus of ND- or KD-fed mice after KA treatment. (A) Western blot showing phosphorylation of hippocampal AMPK. (B) Quantification of hippocampal AMPK activation (ratio of p-AMPK to total AMPK) based on Western blot analysis. (C) Western blot showing phosphorylation of hippocampal ACC. (D) Quantification of hippocampal ACC inhibition based on Western blot analysis. Densitometry values of p-AMPK or p-ACC were normalized to AMPK or ACC and represented as a ratio, respectively. This experiment was repeated independently three times. Data (n = 6) are presented as the mean ± S.E.M. a Significantly different from ND; b significantly different from ND+KA; c significantly different from KD+KA; d significantly different from KD. Significance = P < 0.05 by ANOVA.
(Fig. 4). ND-fed KA-treated mice showed an increase in HSP70 expression in the hippocampus (P < 0.05). However, KD attenuated the increase of HSP70 expression in the hippocampus of KA-treated mice (P < 0.05). In contrast to the observed increase in AMPK and ACC expression in ND-fed mice, there was no increase in hippocampal HSP expression in KD-fed mice. It is widely accepted that the overall function of the AMPK cascade is to control metabolism in response to variations in the energy status of the cell. AMPK is not only expressed in the brain during energy-deprived states, such as fasting and excitotoxicity, but is also induced by consumption of the KD. In addition, AMPK was recently suggested to effect survival and apoptosis [13,21], although its primary function is to mediate cellular adaptation to metabolic stresses. In this study, we have demonstrated that KD decreases activation of the AMPK/ACC cascade and expression of HSP70 in the
Fig. 4. Effect of KD on HSP70 expression in the hippocampus of ND- or KD-fed mice after KA treatment. (A) Western blot showing HSP70 expression in the hippocampus after KA treatment. (B) Quantitation of hippocampal HSP70 based on Western blot analysis. Densitometry value of HSP70 was normalized to ␣-tubulin and represented as arbitrary units (A.U.). This experiment was repeated independently three times. Data (n = 6) are presented as the mean ± S.E.M. a Significantly different from ND; b significantly different from ND+KA; c significantly different from KD+KA; d significantly different from KD. Significance = P < 0.05 by ANOVA.
hippocampus of KA-treated mice, and protects against hippocampal cell death and KA-induced oxidative stress. Consistent with the results here, a previous study demonstrated that a single injection of KA (30 mg/kg) into ICR mice led to neuronal cell death in the hippocampus [22,24]. We observed numerous pyknotic cells or FJB-positive cells in the hippocampal CA3 region, 48 h after KA treatment. In KD-fed mice, however, KA-induced hippocampal cell death was less than that in ND-fed mice. These results suggest that consumption of a KD before KA treatment may have neuroprotective effects in the hippocampus against KA-induced neuronal cell death. AMPK is a metabolic stress sensor/effector that is activated under conditions of nutrient starvation, oxidative stress, vigorous exercise, or heat shock [4,20,28]. Once activated by metabolic stress, the AMPK cascade seems to monitor cellular energy status and initiate appropriate energy-conserving mechanisms in response to ATP depletion, and is activated by low glucose conditions [13]. AMPK is abundantly expressed in neurons and astrocytes, as well as in peripheral tissues [5], but little is known about its role in the brain. Reports have been contradictory about the effect of AMPK in anti-apoptotic activity or cell death. Long-term stimulation of AMPK with 5-aminoimidazole-4-carboxamide ribonucleoside (AICAR) prevented apoptosis in astrocytes [2]. In contrast, several reports have shown that AMPK promotes cell death in a variety of cell types, including neurons, and that AMPK inhibition protects cells from ATP depletion-induced cell death following exposure to various stressors [12,16,19]. Li et al. demonstrated a deleterious effect of AMPK activation in AMPK ␣-2-deficient mice in which significantly smaller infarct volumes were observed when compared with wild-type littermates [17]. In the present study, we found that the KD decreased KA-induced AMPK activation, consistent with results showing a deleterious effect of AMPK activation. Since a KD results in a tendency for modest activation of AMPK only in non-KA treated KD-fed mice, we could not exclude the possibility that KD decreases the high activation of AMPK induced by KA treatment. The KD is high in fat and low in carbohydrate, and provides insufficient amounts of carbohydrates to meet the metabolic demands of the body. Thus, it mimics the metabolic conditions of fasting,
B.T. Jeon et al. / Neuroscience Letters 453 (2009) 49–53
which results in AMPK activation [1]. Furthermore, Kennedy et al. [14] reported that AMPK activity increased in the liver of animals fed a KD, with a corresponding decrease in ACC activity. AMPK activation can generate ATP via various catabolic pathways, including fatty acid oxidation by phosphorylation of ACC, the enzyme that produces malonyl-CoA from acetyl-CoA in fatty acid biosynthesis [3,9]. In contrast to the metabolism in high-fat diet-fed animals, in KDfed animals, lipid and cholesterol synthesis pathways are reduced and ketogenic pathways are increased, leading to an enhanced flow of excess acetyl-CoA into ketone production [13]. Consistent with increased AMPK expression, ACC activity was decreased in muscle and liver [14]. This could be explained by an increase in the inactive phosphorylated form of ACC. After animals receive a KD, large amounts of acetyl-CoA are generated, leading to the synthesis in the liver of three ketone body species: -hydroxybutyrate, acetoacetate, and acetone [10]. These metabolites enter the brain through the blood–brain barrier (BBB). Under normal diet conditions, utilization of ketone bodies in adult brain is minimal. However, ketone bodies partly replace glucose as fuel for the brain during KD consumption. Theses ketone bodies are converted to acetyl-CoA by several enzymes and then enter the Krebs cycle within mitochondria, leading to the production of ATP. Our results indicate that ACC phosphorylation in the hippocampus was increased after KA treatment, consistent with the stimulatory effect of KA on AMPK activity. This the KA-induced increase in ACC phosphorylation was decreased by the KD. In particular, ACC phosphorylation in KD-fed mice without KA treatment was increased compared to that of ND-fed mice. This result suggests that long-term treatment of KD contributes to increased fatty acid oxidation and ketogenesis, protecting against KA-induced higher energy demands. HSPs contribute to the cellular repair processes via facilitating re-folding of denatured proteins. In KA-treated animal models, intense HSP70 immunofluorescence staining was observed 2 h after KA treatment in rat hippocampal CA3 pyramidal neurons, and then dramatically declined by day 5 [29]. Consistent with other studies, our result shows a significant increase in HSP70 expression in the hippocampus of ND-fed mice 2 days after KA treatment. The KD decreased the increased level of HSP70 in the hippocampus of KAtreated mice. This result indicates that HSP70 serves as indicator of stressed neurons during cell death, and that KD itself does not induce stress. In conclusion, our results confirm that consumption of a KD protects against KA-induced hippocampal cell death through down-regulation of the AMPK cascade. This study suggests that ACC phosphorylation induced by long term KD treatment may be involved in neuroprotection. Furthermore, targeting the AMPK/ACC pathway may provide a novel approach for the treatment of seizures and neurodegenerative disorders like Parkinson’s disease, stroke, and Alzheimer’s disease. Acknowledgements This work was supported by the MRC program of MOST/KOSEF (R13-2005-012-01001-0) and was partially supported by a grant from the Korean Ministry of Education (KRF-2006-005-J04201). References [1] E.E. Bailey, H.H. Pfeifer, E.A. Thiele, The use of diet in the treatment of epilepsy, Epilepsy Behav. 6 (2005) 4–8. [2] C. Blázquez, M.J. Geelen, G. Velasco, M. Guzmán, The AMP-activated protein kinase prevents ceramide synthesis de novo and apoptosis in astrocytes, FEBS Lett. 489 (2001) 149–153.
53
[3] D. Carling, The AMP-activated protein kinase cascade—a unifying system for energy control, Trends Biochem. Sci. 29 (2004) 18–24. [4] J.M. Corton, J.G. Gillespie, D.G. Hardie, Role of the AMP-activated protein kinase in the cellular stress response, Curr. Biol. 4 (1994) 315–324. [5] C. Culmsee, J. Monnig, B.E. Kemp, M.P. Mattson, AMP-activated protein kinase is highly expressed in neurons in the developing rat brain and promotes neuronal survival following glucose deprivation, J. Mol. Neurosci. 17 (2001) 45–58. [6] Y. Dagon, Y. Avraham, E.M. Berry, AMPK activation regulates apoptosis, adipogenesis, and lipolysis by eIF2alpha in adipocytes, Biochem. Biophys. Res. Commun. 340 (2006) 43–47. [7] D.C. DeVivo, M.P. Leckie, J.S. Ferrendelli, D.B. McDougal, Chronic ketosis and cerebral metabolism, Ann. Neurol. 3 (1978) 331–337. [8] C. Escartin, K. Pierre, A. Colin, E. Brouillet, T. Delzescaux, M. Guillermier, M. Dhenain, N. Déglon, P. Hantraye, L. Pellerin, G. Bonvento, Activation of astrocytes by CNTF induces metabolic plasticity and increases resistance to metabolic insults, J. Neurosci. 27 (2007) 7094–7104. [9] D.G. Hardie, A. Pan, Regulation of fatty acid synthesis and oxidation by the AMP-activated protein kinase, Biochem. Soc. Trans. 30 (2002) 1064–1070. [10] A.L. Hartman, M. Gasior, E.P. Vining, M.A. Rogawski, The neuropharmacology of the ketogenic diet, Pediatr. Neurol. 36 (2007) 281–292. [11] R.Q. Hu, S. Koh, T. Torgerson, A.J. Cole, Neuronal stress and injury in C57/BL mice after systemic kainic acid administration, Brain Res. 810 (1998) 229–240. [12] J.E. Jung, J. Lee, J. Ha, S.S. Kim, Y.H. Cho, H.H. Baik, I. Kang, 5-Aminoimidazole4-carboxamide-ribonucleoside enhances oxidative stress-induced apoptosis through activation of nuclear factor-kappaB in mouse Neuro 2a neuroblastoma cells, Neurosci. Lett. 354 (2004) 197–200. [13] B.E. Kemp, K.I. Mitchelhill, D. Stapleton, B.J. Michell, Z.P. Chen, L.A. Witters, Dealing with energy demand: the AMP-activated protein kinase, Trends Biochem. Sci. 24 (1999) 22–25. [14] A.R. Kennedy, P. Pissios,.H. Out, B. Xue, K. Asakura, N. Furukawa, F.E. Marino, F.F. Liu, B.B. Kahn, T.A. Libermann, E. Maratos-Flier, R. Roberson, A high-fat, ketogenic diet induces a unique metabolic state in mice, Am. J. Physiol. Endocrinol. Metab. 292 (2007) E1724–1739. [15] T. Kondo, F.R. Sharp, J. Honkaniemi, S. Mikawa, C.J. Epstein, P.H. Chan, DNA fragmentation and Prolonged expression of c-fos, c-jun, and hsp70 in kainic acid-induced neuronal cell death in transgenic mice overexpressing human CuZn-superoxide dismutase, J. Cereb. Blood Flow Metab. 17 (1997) 241–256. [16] L.E. Landree, A.L. Hanlon, D.W. Strong, G. Rumbaugh, I.M. Miller, J.N. Thupari, E.C. Connolly, R.L. Huganir, C. Richardson, L.A. Witters, F.P. Kuhajda, G.V. Ronnett, C75, a fatty acid synthase inhibitor, modulates AMP-activated protein kinase to alter neuronal energy metabolism, J. Biol. Chem. 279 (2004) 3817–3827. [17] J. Li, Z. Zeng, B. Viollet, G.V. Ronnett, L.D. McCullough, Neuroprotective effects of adenosine monophosphate-activated protein kinase inhibition and gene deletion in stroke, Stroke 38 (2007) 2992–2999. [18] S.A. Masino, J.D. Geiger, Are purines mediators of the anticonvulsant/neuroprotective effects of ketogenic diets? Trends Neurosci. 31 (2008) 273–278. [19] L.D. McCullough, Z. Zeng, H. Li, L.E. Landree, J. McFadden, G.V. Ronnett, Pharmacological inhibition of AMP-activated protein kinase provides neuroprotection in stroke, J. Biol. Chem. 280 (2005) 20493–20502. [20] F. Moore, J. Weekes, D.G. Hardie, Evidence that AMP triggers phosphorylation as well as direct allosteric activation of rat liver AMP-activated protein kinase. A sensitive mechanism to protect the cell against ATP depletion, Eur. J. Biochem. 199 (1991) 691–697. [21] Y. Nakatsu, Y. Kotake, A. Hino, S. Ohta, Activation of AMP-activated protein kinase by tributyltin induces neuronal cell death, Toxicol. Appl. Pharmacol. 230 (2008) 358–363. [22] H.S. Noh, Y.S. Kim, H.P. Lee, K.M. Chung, D.W. Kim, S.S. Kang, G.J. Cho, W.S. Choi, The protective effect of a ketogenic diet on kainic acid-induced hippocampal cell death in the male ICR mice, Epilepsy Res. 53 (2003) 119–128. [23] S. Ramamurthy, G.V. Ronnett, Developing a head for energy sensing: AMPactivated protein kinase as a multifunctional metabolic sensor in the brain, J. Physiol. 574 (2006) 85–93. [24] H.J. Shin, J.Y. Lee, E. Son, D.H. Lee, H.J. Kim, S.S. Kang, G.J. Cho, W.S. Choi, G.S. Roh, Curcumin attenuates the kainic acid-induced hippocampal cell death in the mice, Neurosci. Lett. 416 (2007) 49–54. [25] T.D. Swink, E.P. Vining, J.M. Freeman, The ketogenic diet: 1997, Adv. Pediatr. 44 (1997) 297–329. [26] M.T. Todorova, P. Tandon, R.A. Madore, C.E. Stafstrom, T.N. Seyfried, The ketogenic diet inhibits epileptogenesis in EL mice: a genetic model for idiopathic epilepsy, Epilepsia 41 (2000) 933–940. [27] K. Vass, M.L. Berger, T.S. Nowak Jr., W.J. Welch, H. Lassmann, Induction of stress protein HSP70 in nerve cells after status epilepticus in the rat, Neurosci. Lett. 100 (1989) 259–264. [28] D. Vavvas, A. Apazidis, A.K. Saha, J. Gamble, A. Patel, B.E. Kemp, L.A. Witters, N.B. Ruderman, Contraction-induced changes in acetyl-CoA carboxylase and 5 AMP-activated kinase in skeletal muscle, J. Biol. Chem. 272 (1997) 13255–13261. [29] T. Yang, C. Hsu, W. Liao, J.S. Chuang, Heat shock protein 70 expression in epilepsy suggests stress rather than protection, Acta Neuropathol. 115 (2008) 219–230. [30] M. Yudkoff, Y. Daikhin, I. Nissim, A. Lazarow, I. Nissim, Ketogenic diet, amino acid metabolism, and seizure control, J. Neurosci. Res. 66 (2001) 931–940.
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Ketogenic diet attenuates kainic acid-induced hippocampal cell death by decreasing AMPK/ACC pathway activity and HSP70 Byeong Tak Jeon a , Dong Hoon Lee a , Kyu Hong Kim b , Hyun Joon Kim a , Sang Soo Kang a , Gyeong Jae Cho a , Wan Sung Choi a , Gu Seob Roh a,∗ a
Department of Anatomy and Neurobiology, Institute of Health Sciences, Medical Research Center for Neural Dysfunction, Gyeongsang National University, School of Medicine, 92 Chilam-dong, Jinju, Gyeongnam, 660-751, Republic of Korea Department of Neurosurgery, Masan Samgsung Hospital, Sungkyunkwan University School of Medicine, Masan, Gyeongnam, 630-723, Republic of Korea
b
a r t i c l e
i n f o
Article history: Received 11 November 2008 Received in revised form 28 January 2009 Accepted 28 January 2009 Keywords: Ketogenic diet Kainic acid AMPK ACC HSP70 Hippocampus
a b s t r a c t The ketogenic diet (KD) prevents kainic acid (KA)-induced hippocampal cell death. There are reports that AMP-activated protein kinase (AMPK) activation regulates the intracellular signaling pathways involved in cellular survival or apoptotic cell death. In this study, we investigated the effect of the KD consumption on the expression of signaling pathway proteins AMPK and ACC, and heat shock protein (HSP) 70 in mouse hippocampus after KA treatment. Mice were fed the KD for 6 weeks and then sacrificed 48 h after KA (30 mg/kg) injection. The marked cell death found commonly in normal diet (ND)-fed mice treated with KA was not observed in the KD-fed KA-treated mice. Western blot analysis revealed that phosphorylation of AMPK and ACC was increased after KA treatment. However, phosphorylation of these proteins was reduced in those animals that received the KD. In addition, increased expression of HSP70 in the hippocampus of KA-treated mice was decreased in animals receiving the KD. These results indicate that the KD promotes neuroprotective effects through suppression of the AMPK cascade and that HSP70 is involved in neuronal cell death or oxidative stress. © 2009 Elsevier Ireland Ltd. All rights reserved.
The ketogenic diet (KD) is a high-fat and low-carbohydrate diet primarily used to treat juvenile refractory epilepsy [1,25]. Despite its use in clinical therapy and experimental animal studies, it remains unclear what mechanisms underlie its seizure-suppressive action and neuroprotection. Like fasting, the KD has beneficial effects on epilepsy, results in limited glucose availability due to a very low carbohydrate intake, and forces the body to use ketones as an alternate source of acetyl-CoA to generate ATP [26,30]. During consumption of the KD, marked alterations in brain energy metabolism occur. Ketones partly replace glucose as fuel, increasing levels of ATP and purines which are critically involved in neuron–glia interactions, neuromodulation, and synaptic plasticity [10,18]. AMP-activated protein kinase (AMPK) acts as a multifunctional metabolic sensor in the brain [13,23]. AMPK is activated under conditions of glucose deprivation, heat shock, oxidative stress, and ischemia [4]. Once activated, AMPK suppresses the key enzymes involved in ATP-consuming anabolic pathways and increases cellular ATP supply [13]. AMPK stimulates fatty acid oxidation by phosphorylating and inhibiting acetyl-CoA carboxylase (ACC) [9]. AMPK regulates a variety of neuronal cell functions including
∗ Corresponding author. Tel.: +82 55 751 8735; fax: +82 55 759 0779. E-mail address: [email protected] (G.S. Roh). 0304-3940/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2009.01.068
survival, apoptosis, and brain metabolic plasticity [2,5,8]. Modest activation of AMPK induces neurogenesis and improved cognition in animals, but over-activation reduces cognition and increases neural apoptosis and mortality [6]. Under conditions of oxidative stress, heat shock protein (HSP) expression temporarily increases to assist recovery by enhancing the cell’s ability to cope with increased levels of denatured proteins [29]. HSP70, which is induced in areas damaged by KA-induced seizure, is a biochemical marker of neural injury and influences apoptotic cell death [15,27]. Neuronal cells need large amounts of energy, and abnormalities in energy metabolism are detrimental to neurophysiological functions. Previous reports have suggested that the KD increases ATP and the phosphocreatine-to-creatine ratio of energy reserves [7]. Since a seizure induces massive energy depletion and is followed by neuronal cell death, we hypothesized that preactivation of AMPK may itself play an important role in energy metabolism after seizure activity. The present study was undertaken to evaluate the expression of signal pathway proteins AMPK and ACC, and HSP70 in KA-induced hippocampal cell death following administration of the KD. Male ICR mice purchased from Japan SLC (Hamamatsu, Japan) were maintained in the animal facility of the Gyeongsang National University School of Medicine. Mice were treated in accordance
50
B.T. Jeon et al. / Neuroscience Letters 453 (2009) 49–53
with standard guidelines for laboratory animal care. Mice were fed ad libitum either a Harlan Teklad TD 96355 diet (KD group) or a standard rodent chow (normal diet, ND group) for 6 weeks, beginning at postnatal day 21. KD-fed mice were fasted overnight prior to initiation of the KD to facilitate the development of ketonemia [25]. After 6 weeks consuming a KD or ND, mice were treated with a subcutaneous injection of 30 mg/kg KA (Ascent Scientific, North Somerset, UK) emulsified in 0.9% normal saline. All animals were seizurenaïve when tested. Mice in control groups received 0.9% normal saline. Seizure behavior was monitored for 2 h after KA injection, and seizures were categorized into five grades (I–V), according to a previously defined scale [11]. Each experimental group (n = 10) was sacrificed at 48 h after KA treatment. For tissue analysis, mice (n = 4 per group) were perfused transcardially with heparinized saline followed by 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS). After 6 h post-fixation in the same fixative, the brains were sequentially immersed in 0.1 M PBS containing 15% sucrose and then in PBS containing 30% sucrose at 4 ◦ C, until they were completely submerged. The brains were cut into 40 m coronal sections and stained with cresyl violet, and the slides were evaluated by microscopy. Neuronal loss in the pyramidal layers of each section was scored on a 0–3 scale by observers blinded to the treatment conditions: 0, no lesion; 1, minimal lesion localized to CA3; 2, cellular loss in CA3 with some preservation of cellular architecture and some normal cellular components; 3, complete disruption of normal cellular architecture and components. To visualize degenerating neurons by Fluoro-Jade B (FJB) staining, slides containing brain sections were air-dried for 30 min, and then immersed in 1% sodium hydroxide in 80% ethanol for 5 min. The slides were then transferred to 70% ethanol and incubated for 2 min before transfer to deionized water (DW) for 2 min incubation, after which they were incubated in freshly prepared 0.06% KMnO4 for 10 min. Slides were then rinsed in DW for another 2 min before being incubated in 0.0004% FJB solution (Chemicon, CA, USA) for 20 min. The slides were rinsed three times in DW for 1 min, and excess water was removed by briefly hanging the slides vertically on a paper towel. They were then air dried for 10 min on a warmer, and immersed in xylene for 2 min before placing a coverslip with permount. The degenerating neurons were visualized as glowing green cells using a fluorescence microscope. The number of FJB-stained degenerating neurons was counted manually. For protein extraction, frozen mouse hippocampal samples (n = 6 per group) were transferred to sterile 1.5 mL microcentrifuge tubes containing 200 L lysis buffer (15 mM HEPES, pH 7.9, 0.25 M sucrose, 60 mM KCl, 10 mM NaCl, 1 mM EGTA, 1 mM PMSF, and 2 mM NaF). Homogenized tissues were incubated for 10 min on ice and then sonicated. Samples were then centrifuged at 4 ◦ C for 30 min at 12,000 rpm, and the supernatants were transferred to clean vials. Protein concentrations were determined using a Bio-Rad protein assay (Bio-Rad, USA), and samples were stored at −80 ◦ C. For phospho-AMPK (p-AMPK), AMPK, phospho-ACC (p-ACC), ACC, and HSP70 protein analysis, hippocampal lysates (30 g per lane) were separated by 8 or 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by electrophoretic transfer onto a polyvinylidene difluoride membrane (Millipore, MA, USA). The membranes were probed with antibodies against p-AMPK (Cell Signaling Technology, MA, USA), AMPK (Cell Signaling), p-ACC (Cell Signaling), ACC (Cell Signaling), or HSP70 (Santa Cruz Biotechnology, CA) and visualized using an ECL substrate (Pierce, Rockford, IL). To determine the neuroprotective effects of KD, median histopathologic score values were analyzed by nonparametric test (Mann–Whitney rank sum test). Differences between the groups (ND, ND+KA, KD+KA, and KD) were determined by one-way analysis of variance (ANOVA), followed by post hoc analysis using a Student–Newman–Keuls test. Values are expressed as the mean ± standard error of the mean (S.E.M.). A P value < 0.05 was considered statistically significant.
Fig. 1. Effect of KD on body weight and latency to seizure. (A) Ketogenic diet was begun at postnatal day 21 and was continued for 6 weeks. To promote ketosis induction, KD-fed mice group was deprived of food and water overnight prior to diet initiation. During the dietary treatment, body weight of ND-fed mice (n = 20) and KDfed mice (n = 20) showed no significant differences. (B) All mice were treated with KA (30 mg/kg, s.c.). In KD-fed mice, the latency to seizure onset was delayed compared to ND-fed mice. The data represent the mean ± S.E.M. of 20 mice. Significant (P < 0.05) difference compared with ND-fed mice.
KD-fed mice tolerated their diet well for 6 weeks. Other than the oily appearance of their fur, KD-fed mice showed no difference in behavior or health when compared with ND-fed mice. Apart from an initial weight loss in the ND- or KD-fed mice, mice in both groups experienced a similar weight gain within the time allotted (Fig. 1A). ND- and KD-fed mice showed increasing immobility that resulted in rigidity within 7 and 10 min (P = 0.01), respectively, after KA (30 mg/kg) injection (Fig. 1B). Although the latency of the seizures elicited by KA was retarded in KD-fed mice, the severity of the seizures was not significantly attenuated. During the 2 h post-injection observation period, all of the mice in both groups displayed head bobbing, circling behavior, and falling behavior, corresponding to grade IV. To determine whether KD plays a protective role in KA-induced hippocampal cell death, we performed cresyl violet and FJB staining (Fig. 2) of brain tissue sections. Cresyl violet staining revealed significant cell loss in the CA3 pyramidal layers of ND-fed KA-induced seizure mice, with pyramidal neurons appearing pyknotic (Fig. 2B). In contrast, cresyl violet staining in KD-fed KA-treated mice showed a significant level of protection against hippocampal cell death (Fig. 2C). Neuronal loss in ND+KA-fed mice (median score = 2.5) was significantly higher (P < 0.001) than that observed in KD+KA-fed mice (median score = 0.50). The FJB finding was consistent with cresyl violet staining (Fig. 2E and F). FJB-positive cells were observed in the CA3 region of ND-fed KA-induced seizure mice, whereas the signal was markedly reduced in KA-injected KD-fed mice (ND+KA-treated mice: 55.0 ± 2.74, KD+KA-treated mice: 7.5 ± 0.65; P < 0.001). No FJB-positive cells were observed in non-KA injected ND- or KD-fed mice.
B.T. Jeon et al. / Neuroscience Letters 453 (2009) 49–53
51
Fig. 2. Effect of KD on neuroprotection in the hippocampus of ND- or KD-fed mice after KA treatment. (A)–(D) Cresyl violet stained sections, showing the cytoarchitecture from (A) ND-, (B) ND+KA-, (C) KD+KA-, and (D) KD-treated mice. (E) and (F) Fluoro-Jade B stained section, showing the cytoarchitecture from (E) ND+KA- and (F) KD+KA-treated mice. Magnification: 400×. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
To explore whether phosphorylation of AMPK or ACC in the hippocampus of KA-treated mice is altered by KD, we performed Western blot analysis (Fig. 3). KA treatment resulted in an increased level of p-AMPK (P < 0.05) (Fig. 3A and B). However, KD attenuated the increase of AMPK activation in the hippocampus of KA-treated mice (P = 0.45). KD-fed mice showed a tendency toward increased AMPK activation over that of ND-fed mice; however, this
increase was not statistically significant (P = 0.059). In addition, a KA-induced increase in ACC phosphorylation was significantly inhibited by the KD (P < 0.05) (Fig. 3C and D). Consistent with results of AMPK activation, there was an increase in ACC phosphorylation in the hippocampus of KD-fed mice (P < 0.05). Finally, Western blot analysis was performed to determine the effect of KD on HSP70 expression under conditions of KA-induced hippocampal cell death
52
B.T. Jeon et al. / Neuroscience Letters 453 (2009) 49–53
Fig. 3. Effect of KD on the phosphorylation of AMPK and ACC in the hippocampus of ND- or KD-fed mice after KA treatment. (A) Western blot showing phosphorylation of hippocampal AMPK. (B) Quantification of hippocampal AMPK activation (ratio of p-AMPK to total AMPK) based on Western blot analysis. (C) Western blot showing phosphorylation of hippocampal ACC. (D) Quantification of hippocampal ACC inhibition based on Western blot analysis. Densitometry values of p-AMPK or p-ACC were normalized to AMPK or ACC and represented as a ratio, respectively. This experiment was repeated independently three times. Data (n = 6) are presented as the mean ± S.E.M. a Significantly different from ND; b significantly different from ND+KA; c significantly different from KD+KA; d significantly different from KD. Significance = P < 0.05 by ANOVA.
(Fig. 4). ND-fed KA-treated mice showed an increase in HSP70 expression in the hippocampus (P < 0.05). However, KD attenuated the increase of HSP70 expression in the hippocampus of KA-treated mice (P < 0.05). In contrast to the observed increase in AMPK and ACC expression in ND-fed mice, there was no increase in hippocampal HSP expression in KD-fed mice. It is widely accepted that the overall function of the AMPK cascade is to control metabolism in response to variations in the energy status of the cell. AMPK is not only expressed in the brain during energy-deprived states, such as fasting and excitotoxicity, but is also induced by consumption of the KD. In addition, AMPK was recently suggested to effect survival and apoptosis [13,21], although its primary function is to mediate cellular adaptation to metabolic stresses. In this study, we have demonstrated that KD decreases activation of the AMPK/ACC cascade and expression of HSP70 in the
Fig. 4. Effect of KD on HSP70 expression in the hippocampus of ND- or KD-fed mice after KA treatment. (A) Western blot showing HSP70 expression in the hippocampus after KA treatment. (B) Quantitation of hippocampal HSP70 based on Western blot analysis. Densitometry value of HSP70 was normalized to ␣-tubulin and represented as arbitrary units (A.U.). This experiment was repeated independently three times. Data (n = 6) are presented as the mean ± S.E.M. a Significantly different from ND; b significantly different from ND+KA; c significantly different from KD+KA; d significantly different from KD. Significance = P < 0.05 by ANOVA.
hippocampus of KA-treated mice, and protects against hippocampal cell death and KA-induced oxidative stress. Consistent with the results here, a previous study demonstrated that a single injection of KA (30 mg/kg) into ICR mice led to neuronal cell death in the hippocampus [22,24]. We observed numerous pyknotic cells or FJB-positive cells in the hippocampal CA3 region, 48 h after KA treatment. In KD-fed mice, however, KA-induced hippocampal cell death was less than that in ND-fed mice. These results suggest that consumption of a KD before KA treatment may have neuroprotective effects in the hippocampus against KA-induced neuronal cell death. AMPK is a metabolic stress sensor/effector that is activated under conditions of nutrient starvation, oxidative stress, vigorous exercise, or heat shock [4,20,28]. Once activated by metabolic stress, the AMPK cascade seems to monitor cellular energy status and initiate appropriate energy-conserving mechanisms in response to ATP depletion, and is activated by low glucose conditions [13]. AMPK is abundantly expressed in neurons and astrocytes, as well as in peripheral tissues [5], but little is known about its role in the brain. Reports have been contradictory about the effect of AMPK in anti-apoptotic activity or cell death. Long-term stimulation of AMPK with 5-aminoimidazole-4-carboxamide ribonucleoside (AICAR) prevented apoptosis in astrocytes [2]. In contrast, several reports have shown that AMPK promotes cell death in a variety of cell types, including neurons, and that AMPK inhibition protects cells from ATP depletion-induced cell death following exposure to various stressors [12,16,19]. Li et al. demonstrated a deleterious effect of AMPK activation in AMPK ␣-2-deficient mice in which significantly smaller infarct volumes were observed when compared with wild-type littermates [17]. In the present study, we found that the KD decreased KA-induced AMPK activation, consistent with results showing a deleterious effect of AMPK activation. Since a KD results in a tendency for modest activation of AMPK only in non-KA treated KD-fed mice, we could not exclude the possibility that KD decreases the high activation of AMPK induced by KA treatment. The KD is high in fat and low in carbohydrate, and provides insufficient amounts of carbohydrates to meet the metabolic demands of the body. Thus, it mimics the metabolic conditions of fasting,
B.T. Jeon et al. / Neuroscience Letters 453 (2009) 49–53
which results in AMPK activation [1]. Furthermore, Kennedy et al. [14] reported that AMPK activity increased in the liver of animals fed a KD, with a corresponding decrease in ACC activity. AMPK activation can generate ATP via various catabolic pathways, including fatty acid oxidation by phosphorylation of ACC, the enzyme that produces malonyl-CoA from acetyl-CoA in fatty acid biosynthesis [3,9]. In contrast to the metabolism in high-fat diet-fed animals, in KDfed animals, lipid and cholesterol synthesis pathways are reduced and ketogenic pathways are increased, leading to an enhanced flow of excess acetyl-CoA into ketone production [13]. Consistent with increased AMPK expression, ACC activity was decreased in muscle and liver [14]. This could be explained by an increase in the inactive phosphorylated form of ACC. After animals receive a KD, large amounts of acetyl-CoA are generated, leading to the synthesis in the liver of three ketone body species: -hydroxybutyrate, acetoacetate, and acetone [10]. These metabolites enter the brain through the blood–brain barrier (BBB). Under normal diet conditions, utilization of ketone bodies in adult brain is minimal. However, ketone bodies partly replace glucose as fuel for the brain during KD consumption. Theses ketone bodies are converted to acetyl-CoA by several enzymes and then enter the Krebs cycle within mitochondria, leading to the production of ATP. Our results indicate that ACC phosphorylation in the hippocampus was increased after KA treatment, consistent with the stimulatory effect of KA on AMPK activity. This the KA-induced increase in ACC phosphorylation was decreased by the KD. In particular, ACC phosphorylation in KD-fed mice without KA treatment was increased compared to that of ND-fed mice. This result suggests that long-term treatment of KD contributes to increased fatty acid oxidation and ketogenesis, protecting against KA-induced higher energy demands. HSPs contribute to the cellular repair processes via facilitating re-folding of denatured proteins. In KA-treated animal models, intense HSP70 immunofluorescence staining was observed 2 h after KA treatment in rat hippocampal CA3 pyramidal neurons, and then dramatically declined by day 5 [29]. Consistent with other studies, our result shows a significant increase in HSP70 expression in the hippocampus of ND-fed mice 2 days after KA treatment. The KD decreased the increased level of HSP70 in the hippocampus of KAtreated mice. This result indicates that HSP70 serves as indicator of stressed neurons during cell death, and that KD itself does not induce stress. In conclusion, our results confirm that consumption of a KD protects against KA-induced hippocampal cell death through down-regulation of the AMPK cascade. This study suggests that ACC phosphorylation induced by long term KD treatment may be involved in neuroprotection. Furthermore, targeting the AMPK/ACC pathway may provide a novel approach for the treatment of seizures and neurodegenerative disorders like Parkinson’s disease, stroke, and Alzheimer’s disease. Acknowledgements This work was supported by the MRC program of MOST/KOSEF (R13-2005-012-01001-0) and was partially supported by a grant from the Korean Ministry of Education (KRF-2006-005-J04201). References [1] E.E. Bailey, H.H. Pfeifer, E.A. Thiele, The use of diet in the treatment of epilepsy, Epilepsy Behav. 6 (2005) 4–8. [2] C. Blázquez, M.J. Geelen, G. Velasco, M. Guzmán, The AMP-activated protein kinase prevents ceramide synthesis de novo and apoptosis in astrocytes, FEBS Lett. 489 (2001) 149–153.
53
[3] D. Carling, The AMP-activated protein kinase cascade—a unifying system for energy control, Trends Biochem. Sci. 29 (2004) 18–24. [4] J.M. Corton, J.G. Gillespie, D.G. Hardie, Role of the AMP-activated protein kinase in the cellular stress response, Curr. Biol. 4 (1994) 315–324. [5] C. Culmsee, J. Monnig, B.E. Kemp, M.P. Mattson, AMP-activated protein kinase is highly expressed in neurons in the developing rat brain and promotes neuronal survival following glucose deprivation, J. Mol. Neurosci. 17 (2001) 45–58. [6] Y. Dagon, Y. Avraham, E.M. Berry, AMPK activation regulates apoptosis, adipogenesis, and lipolysis by eIF2alpha in adipocytes, Biochem. Biophys. Res. Commun. 340 (2006) 43–47. [7] D.C. DeVivo, M.P. Leckie, J.S. Ferrendelli, D.B. McDougal, Chronic ketosis and cerebral metabolism, Ann. Neurol. 3 (1978) 331–337. [8] C. Escartin, K. Pierre, A. Colin, E. Brouillet, T. Delzescaux, M. Guillermier, M. Dhenain, N. Déglon, P. Hantraye, L. Pellerin, G. Bonvento, Activation of astrocytes by CNTF induces metabolic plasticity and increases resistance to metabolic insults, J. Neurosci. 27 (2007) 7094–7104. [9] D.G. Hardie, A. Pan, Regulation of fatty acid synthesis and oxidation by the AMP-activated protein kinase, Biochem. Soc. Trans. 30 (2002) 1064–1070. [10] A.L. Hartman, M. Gasior, E.P. Vining, M.A. Rogawski, The neuropharmacology of the ketogenic diet, Pediatr. Neurol. 36 (2007) 281–292. [11] R.Q. Hu, S. Koh, T. Torgerson, A.J. Cole, Neuronal stress and injury in C57/BL mice after systemic kainic acid administration, Brain Res. 810 (1998) 229–240. [12] J.E. Jung, J. Lee, J. Ha, S.S. Kim, Y.H. Cho, H.H. Baik, I. Kang, 5-Aminoimidazole4-carboxamide-ribonucleoside enhances oxidative stress-induced apoptosis through activation of nuclear factor-kappaB in mouse Neuro 2a neuroblastoma cells, Neurosci. Lett. 354 (2004) 197–200. [13] B.E. Kemp, K.I. Mitchelhill, D. Stapleton, B.J. Michell, Z.P. Chen, L.A. Witters, Dealing with energy demand: the AMP-activated protein kinase, Trends Biochem. Sci. 24 (1999) 22–25. [14] A.R. Kennedy, P. Pissios,.H. Out, B. Xue, K. Asakura, N. Furukawa, F.E. Marino, F.F. Liu, B.B. Kahn, T.A. Libermann, E. Maratos-Flier, R. Roberson, A high-fat, ketogenic diet induces a unique metabolic state in mice, Am. J. Physiol. Endocrinol. Metab. 292 (2007) E1724–1739. [15] T. Kondo, F.R. Sharp, J. Honkaniemi, S. Mikawa, C.J. Epstein, P.H. Chan, DNA fragmentation and Prolonged expression of c-fos, c-jun, and hsp70 in kainic acid-induced neuronal cell death in transgenic mice overexpressing human CuZn-superoxide dismutase, J. Cereb. Blood Flow Metab. 17 (1997) 241–256. [16] L.E. Landree, A.L. Hanlon, D.W. Strong, G. Rumbaugh, I.M. Miller, J.N. Thupari, E.C. Connolly, R.L. Huganir, C. Richardson, L.A. Witters, F.P. Kuhajda, G.V. Ronnett, C75, a fatty acid synthase inhibitor, modulates AMP-activated protein kinase to alter neuronal energy metabolism, J. Biol. Chem. 279 (2004) 3817–3827. [17] J. Li, Z. Zeng, B. Viollet, G.V. Ronnett, L.D. McCullough, Neuroprotective effects of adenosine monophosphate-activated protein kinase inhibition and gene deletion in stroke, Stroke 38 (2007) 2992–2999. [18] S.A. Masino, J.D. Geiger, Are purines mediators of the anticonvulsant/neuroprotective effects of ketogenic diets? Trends Neurosci. 31 (2008) 273–278. [19] L.D. McCullough, Z. Zeng, H. Li, L.E. Landree, J. McFadden, G.V. Ronnett, Pharmacological inhibition of AMP-activated protein kinase provides neuroprotection in stroke, J. Biol. Chem. 280 (2005) 20493–20502. [20] F. Moore, J. Weekes, D.G. Hardie, Evidence that AMP triggers phosphorylation as well as direct allosteric activation of rat liver AMP-activated protein kinase. A sensitive mechanism to protect the cell against ATP depletion, Eur. J. Biochem. 199 (1991) 691–697. [21] Y. Nakatsu, Y. Kotake, A. Hino, S. Ohta, Activation of AMP-activated protein kinase by tributyltin induces neuronal cell death, Toxicol. Appl. Pharmacol. 230 (2008) 358–363. [22] H.S. Noh, Y.S. Kim, H.P. Lee, K.M. Chung, D.W. Kim, S.S. Kang, G.J. Cho, W.S. Choi, The protective effect of a ketogenic diet on kainic acid-induced hippocampal cell death in the male ICR mice, Epilepsy Res. 53 (2003) 119–128. [23] S. Ramamurthy, G.V. Ronnett, Developing a head for energy sensing: AMPactivated protein kinase as a multifunctional metabolic sensor in the brain, J. Physiol. 574 (2006) 85–93. [24] H.J. Shin, J.Y. Lee, E. Son, D.H. Lee, H.J. Kim, S.S. Kang, G.J. Cho, W.S. Choi, G.S. Roh, Curcumin attenuates the kainic acid-induced hippocampal cell death in the mice, Neurosci. Lett. 416 (2007) 49–54. [25] T.D. Swink, E.P. Vining, J.M. Freeman, The ketogenic diet: 1997, Adv. Pediatr. 44 (1997) 297–329. [26] M.T. Todorova, P. Tandon, R.A. Madore, C.E. Stafstrom, T.N. Seyfried, The ketogenic diet inhibits epileptogenesis in EL mice: a genetic model for idiopathic epilepsy, Epilepsia 41 (2000) 933–940. [27] K. Vass, M.L. Berger, T.S. Nowak Jr., W.J. Welch, H. Lassmann, Induction of stress protein HSP70 in nerve cells after status epilepticus in the rat, Neurosci. Lett. 100 (1989) 259–264. [28] D. Vavvas, A. Apazidis, A.K. Saha, J. Gamble, A. Patel, B.E. Kemp, L.A. Witters, N.B. Ruderman, Contraction-induced changes in acetyl-CoA carboxylase and 5 AMP-activated kinase in skeletal muscle, J. Biol. Chem. 272 (1997) 13255–13261. [29] T. Yang, C. Hsu, W. Liao, J.S. Chuang, Heat shock protein 70 expression in epilepsy suggests stress rather than protection, Acta Neuropathol. 115 (2008) 219–230. [30] M. Yudkoff, Y. Daikhin, I. Nissim, A. Lazarow, I. Nissim, Ketogenic diet, amino acid metabolism, and seizure control, J. Neurosci. Res. 66 (2001) 931–940.